EP3968594B1

EP3968594B1 - Method, apparatus and system for selecting mobile edge computing node

Info

Publication number: EP3968594B1
Application number: EP20805471.8A
Authority: EP
Inventors: Zhiqiang YOU; Jiajia LOU
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2019-05-10
Filing date: 2020-04-15
Publication date: 2023-11-29
Anticipated expiration: 2040-04-15
Also published as: CN110198307A; US20210273987A1; US11303696B2; JP2022532007A; EP3968594A4; EP3968594A1; KR20210094077A; KR102514250B1; CN110198307B; JP7252356B2; WO2020228469A1

Description

RELATED APPLICATION

This application claims priority to Chinese Patent Application No. 201910387647.7, entitled "METHOD, APPARATUS, AND SYSTEM FOR SELECTING MOBILE EDGE COMPUTING NODE", filed with the China National Intellectual Property Administration on May 10, 2019 .

FIELD OF THE TECHNOLOGY

This application relates to the field of mobile communication, and in particular, to mobile edge computing (MEC) node selection.

BACKGROUND OF THE DISCLOSURE

With the development of science and technology, currently, users have higher and higher requirements for processing and transmission rates of communication services. Particularly, for a fifth-generation mobile communication technology (5G) mobile communication network, requirements for a processing speed are higher. CN 109379774 A discloses an intelligent scheduling method applied to a terminal device, which initiates an access request to an edge node cluster meeting a predetermined condition according to a network connection quality parameter of different edge node clusters; US 2018/0287890 A1 involves provisioning customers of an aggregator, such as a reseller, of a content delivery network; and CN109640319 A relates to a scheduling method based on access information, which can ensure lower latency and higher quality of server for a terminal device. In mobile edge computing, computing, storage, and processing capabilities can be provided by using a nearest mobile access network, to reduce latency. Therefore, a mobile terminal needs to be scheduled to a mobile edge computing (MEC) node closest to the mobile terminal.

SUMMARY

Embodiments of this application provide a method, an apparatus, and a system for selecting an MEC node, to resolve the issue of MEC node selection solution for a 5G mobile communication network.
The invention is set out in the appended set of claims.
The technical solutions provided in the embodiments of this application include at least the following beneficial effects.
By adding an edge cloud gateway, the application is applicable to the conventional DNS addressing process or HTTP DNS requesting process. When the edge cloud gateway obtains a first HTTP service request of a terminal, according to the first HTTP service request carrying an edge-application VIP, the edge cloud gateway can determine a corresponding MEC processing server based on a preset offloading policy, and offload the first HTTP service request to the corresponding MEC processing server. In this way, a mechanism for selecting an MEC node can be achieved by modifying an existing networking procedure without being perceived by the terminal and without changing service logic, and the issue of MEC node selection solution for a 5G mobile communication network is resolved, so that a nearest edge computing node can be selected for purpose of edge acceleration.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic flowchart of a CDN scheduling method in related art.
FIG. 2 is a schematic architectural diagram of a system for selecting an MEC node according to an embodiment of this application.
FIG. 3 is a flowchart of a method for selecting an MEC node according to an embodiment of this application.
FIG. 4 is an interactive flowchart of a method for selecting an MEC node according to an embodiment of this application.
FIG. 5 is an interactive flowchart of another method for selecting an MEC node according to an embodiment of this application.
FIG. 6 is a schematic structural diagram of an apparatus for selecting an MEC node according to an embodiment of this application.
FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of this application.

DESCRIPTION OF EMBODIMENTS

The following clearly and completely describes the technical solutions in embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Apparently, the described embodiments are some of the embodiments of this application rather than all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of this application without creative efforts shall fall within the protection scope of this application.
To facilitate the understanding of the embodiments of this application, the following concepts are briefly introduced.
Content delivery network (CDN): A basic idea of the CDN is to avoid, as many as possible, bottlenecks and links on the Internet that may affect a speed and stability of data transmission, so that content transmission is faster and more stable, and to provide nearest access to the CDN for acceleration.
Domain name system (DNS): As a distributed database that can map a domain name and an Internet Protocol (IP) address to each other, the DNS spares users the trouble of memorizing IP address data strings that can be directly read by a machine, so that it is easier for users to access the Internet.
User plane function (UPF): The UPF is responsible for user plane processing.
Global server load balance (GSLB): The GSLB implements traffic distribution between servers in different regions on a wide area network including the Internet, and directs a user request to the nearest node (or region) to ensure access quality. Mobile edge computing (MEC): The MEC is a technology that deeply integrates the mobile access network with the Internet service based on the 5G evolution architecture. The MEC, by using a wireless access network nearby, provides services and cloud computing functions required by the Internet technology (IT) for telecommunication users, to create a carrier-class service environment with high performance, low latency, and high bandwidth, accelerate rapid download of various content, services, and applications in the network, and allow consumers to experience an uninterrupted high-quality network.
Virtual IP (VIP): The VIP is an IP address that does not link to a specific computer or a network interface controller in a computer. Packets are transmitted to this VIP address, but all data still passes through a real network interface.
Edge-application VIP: Located in an edge equipment room, the edge-application VIP represents a virtual IP address that provides local acceleration for an application.
At present, the gradual development of the 5G mobile communication network leads to higher requirements for processing and transmission rates of communication services. The MEC can provide computing, storage, and processing capabilities by using a nearest mobile access network to reduce latency. However, how to schedule a mobile terminal to an MEC node closest to the mobile terminal is an urgent problem to be resolved. 5G is a cutting-edge technology, where there are relatively few related researches. There is no relevant solution to select an MEC node in a 5G mobile communication network.
In addition, the conventional DNS addressing mechanism in the related art is similar to determining a corresponding IP address for data transmission. FIG. 1 is a schematic flowchart of a CDN scheduling method in related art, the method including the following steps:

Step 1: A terminal transmits a DNS request to a local DNS.
Step 2: The local DNS requests a DNS recursive query from a GSLB.
Step 3: The GSLB returns a best access IP address to the local DNS, and the local DNS caches the IP address.
Step 4: The local DNS transmits the best IP address to the terminal.
Step 5: The terminal transmits a service request to a CDN node.
The CDN node is an edge node, that is, an outer center (OC).
Step 6: The CDN node forwards the service request to an intermediate source CDN node based on an internal route.
Step 7: The intermediate source CDN node forwards the service request to a service source CDN node.
Step 8: The service source CDN node returns requested data to the intermediate source CDN node.
Step 9: The CDN node obtains the data from the intermediate source CDN node and caches the data.
Step 10: The terminal obtains the data from the CDN node.

In the related art, this scheduling method is to locate the terminal by using the IP address allocated by an operator, to schedule the terminal to the nearest CDN node. However, the IP positioning technology has large deviations and low precision, and can only reach the provincial level. However, there are many MEC nodes in this region. Consequently, precision requirements of the MEC nodes cannot be met, that is, the terminal cannot be scheduled to the nearest MEC node.
Therefore, in the embodiments of this application, a method for selecting an MEC node is provided mainly for a 5G mobile communication network, which is mainly based on a conventional DNS addressing mechanism and an HTTP DNS requesting process, modifying an existing networking process without being perceived by an application of a terminal, to achieve access to a nearest edge computing node.
FIG. 2 is an architectural diagram of a system for selecting an MEC node according to an embodiment of this application, which at least includes a UPF 200, an edge cloud gateway 210, a GSLB 220, and an edge controller 230.
It should be noted that in this embodiment of this application, in a network architecture layout, a data channel can connect a central cloud and an edge cloud of a core network of an operator, and to offload data to an edge computing node, the edge cloud gateway 210 and the edge controller 230 are added. The edge cloud gateway 210 is configured on an MEC node side or an edge data center (DC) side, and the edge controller 230 is configured on a central cloud of the core network and communicably connected to the edge cloud gateway 210 and the UPF 200. The GSLB 220 is also deployed on the central cloud side of the core network, and is a device in juxtaposition to the edge controller 230. In addition, the UPF 200, the edge cloud gateway 210, and the GSLB 220 respectively correspond to different pre-configuration rules. The pre-configuration rule represents a routing configuration for service acceleration.

1) The edge controller 230 is mainly responsible for scheduling of global service traffic, and controls the scheduling of service traffic to the edge cloud gateway 210. Then, the edge cloud gateway 210 schedules the service traffic to a specific local MEC processing server, including the following steps:
1. a. Connect to a capability exposure platform of an operator, and pre-configure the UPF 200 and the edge cloud gateway 210 when an edge service is deployed.

Specifically, the edge controller 230 is configured to configure a first pre-configuration rule and a fourth pre-configuration rule of the UPF 200, and a second pre-configuration rule of the edge cloud gateway 210. The foregoing configuration rules are described in detail in the following embodiments.
For example, as shown in FIG. 2, the edge controller 230 may pre-configure the UPF 200 by using the capability exposure platform of the operator and a 5G core network of the operator.

b. Dynamically collect user location information from a 5G core network side of the operator, and select nearest edge computing nodes for respective hosting services, the hosting services representing services that require edge computing in the 5G mobile communication network, that is, services that require service acceleration.
c. Connect to a cloud infrastructure-as-a-service (IaaS) or platform-as-a-service (PaaS) controller to control edge DC resources.

The IaaS and PaaS are not shown in FIG. 2. In practice, the IaaS and PaaS may be located between the edge cloud gateway 210 and the edge controller 230. Substantially, the edge controller 230 can monitor a load status of the edge cloud gateway 210 through the IaaS and PaaS, and control load scheduling of the edge cloud gateway 210.
2) For the GSLB 220, when an edge service is deployed, the GSLB 220 also needs to be pre-configured. Specifically, the GSLB 220 corresponds to a third pre-configuration rule or a fifth pre-configuration rule.
The third pre-configuration rule includes at least:

a. For a DNS request of which a domain name is a preset service acceleration domain name, return a global edge gateway IP address.

Specifically, the GSLB 220 is configured to receive the DNS request transmitted by a terminal, and return the global edge gateway IP address to the terminal in response to determining that the domain name in the DNS request is the preset service acceleration domain name.
The global edge gateway IP address is an IP address that identifies service acceleration, and may be one indicative IP address or a group of indicative IP addresses.
b. For an HTTP DNS request of which a domain name is a preset service acceleration domain name, return a corresponding edge-application VIP based on a source address and the service acceleration domain name in the HTTP DNS request, the source address being an IP address of the edge cloud gateway 210.
For example, when receiving the first HTTP DNS request transmitted by the edge cloud gateway 210, the GSLB 220 returns a first HTTP DNS response to the edge cloud gateway 210. The first HTTP DNS request includes at least a service acceleration domain name and a source address, the source address being the IP address of the edge cloud gateway 210. The first HTTP DNS response includes the edge-application VIP, the edge-application VIP being determined by the GSLB 220 according to the IP address of the edge cloud gateway 210 and the service acceleration domain name in the first HTTP DNS request.
C. For the GSLB 220, alternatively, return configured IP addresses for different source addresses in the first HTTP DNS request by means of manual addition and intervention.
The fifth pre-configuration rule includes at least:

a. For an HTTP DNS request of which a domain name is a preset service acceleration domain name, return a corresponding edge-application VIP based on the service acceleration domain name in the HTTP DNS request.

For example, when receiving the second HTTP DNS request forwarded by the UPF 200, the GSLB 220 returns a second HTTP DNS response, the second HTTP DNS request being forwarded by the UPF 200 upon parsing that a destination address in a second HTTP DNS request transmitted by the terminal is an IP address of the GSLB 220. The second HTTP DNS response includes at least an edge-application VIP, the edge-application VIP being determined by the GSLB 220 according to a service acceleration domain name in the second HTTP DNS request.
That is, the fifth pre-configuration rule of the GSLB 220 is intended for a case in which when the terminal initiates an HTTP DNS request (such as the second HTTP DNS request) for an original domain name, the UPF 200 directly forwards the HTTP DNS request of the terminal to the GSLB 220. In this case, the GSLB 220 obtains the service acceleration domain name therein according to the HTTP DNS request forwarded by the UPF 200, and determines the corresponding edge-application VIP.
b. For the GSLB 220, alternatively, return configured IP addresses for different service acceleration domain names in the second HTTP DNS request by means of manual addition and intervention.
3) For the UPF 200, in this embodiment of this application, the UPF 200 needs to be pre-configured. A pre-configuration function represents a routing configuration for service acceleration. The UPF 200 needs to provide a routing configuration function. The UPF 200 corresponds to the first pre-configuration rule or the fourth pre-configuration rule.
The first pre-configuration rule includes at least:

a. Forward an HTTP service request of which a destination address is the global edge gateway IP address to the edge cloud gateway 210.
For example, the UPF 200 receives a second HTTP service request transmitted by the terminal, and forwards the second HTTP service request to the edge cloud gateway 210 in response to determining that a destination address in the second HTTP service request is the global edge gateway IP address.
b. Forward an HTTP service request of which a destination address is the edge-application VIP to the edge cloud gateway 210.

For example, the UPF 200 receives the first HTTP service request retransmitted by the terminal, and forwards the first HTTP service request to the edge cloud gateway 210 in response to determining that the destination address in the first HTTP service request is the edge-application VIP.
In this embodiment of this application, the first pre-configuration rule of the UPF 200 is intended for a standard DNS requesting process, and in this case, the terminal initiates a DNS request for the original domain name.
The fourth pre-configuration rule includes at least:
forwarding an HTTP service request of which a destination address is the edge-application VIP to the edge cloud gateway 210.
For example, the UPF 200 receives the first HTTP service request transmitted by the terminal, and forwards the first HTTP service request to the edge cloud gateway 210 in response to determining that the destination address of the first HTTP service request is the edge-application VIP.
It should be noted that the fourth pre-configuration rule of the UPF 200 is intended for an HTTP DNS requesting process. In this case, the terminal initiates the second HTTP DNS request for the original domain name. In addition, when the terminal initiates the second HTTP DNS request, the destination address in the second HTTP DNS request is the IP address of the GSLB 220. When the UPF 200 parses that the destination address in the second HTTP DNS request transmitted by the terminal is the IP address of the GSLB, the second HTTP DNS request can be forwarded to the GSLB 220. This rule can be implemented based on an existing forwarding mechanism, and no further configuration is required. Then the UPF 200 receives the second HTTP DNS response returned by the GSLB 220, and forwards the second HTTP DNS response to the terminal, so that the terminal transmits the first HTTP service request based on the edge-application VIP in the second HTTP DNS response. The second HTTP DNS response includes at least the edge-application VIP, the edge-application VIP being determined by the GSLB 220 based on the fifth pre-configuration rule, that is, based on the service acceleration domain name in the second HTTP DNS request.
In this embodiment of this application, the first pre-configuration rule or the fourth pre-configuration rule is configured by the configured edge controller 230, or is configured locally. That is, the UPF 200 not only can be configured and managed by the edge controller 230 in a unified manner, but also can be configured and loaded locally, which is not limited in this embodiment of this application.
It should be further noted that different pre-configuration rule rules correspond to different implementations and cases. The GSLB 220 corresponds to the pre-configuration rules of the UPF 200. For a standard DNS requesting process, if the terminal transmits a DNS request to the GSLB 220, in this case, the GSLB 220 performs a corresponding action based on the third pre-configuration rule, that is, returning the global edge gateway IP address to the terminal. Then the UPF 200 performs a corresponding action based on the first pre-configuration rule, that is, forwarding the second HTTP service request transmitted by the terminal to the edge cloud gateway 210. Then the edge cloud gateway 210 obtains the edge-application VIP from the GSLB 220 and transmits the edge-application VIP to the terminal. The terminal re-initiates the first HTTP service request, then the UPF 200 forwards the first HTTP service request to the edge cloud gateway 210 based on the first pre-configuration rule, and the edge cloud gateway 210 schedules the first HTTP service request to the corresponding MEC processing server.
For an HTTP DNS requesting process, the UPF 200 corresponds to the fourth pre-configuration rule, and the GSLB 220 corresponds to the fifth pre-configuration rule. The terminal initiates the second HTTP DNS request to the GSLB 220 by the UPF 200, then the GSLB 220 returns the edge-application VIP to the terminal based on the corresponding fifth pre-configuration rule, and further, the terminal initiates the first HTTP service request. The UPF 200 performs a corresponding action based on the fourth pre-configuration rule, that it, forwarding the first HTTP service request to the edge cloud gateway 210, and the edge cloud gateway 210 schedules the first HTTP service request to the corresponding MEC processing server. The two implementations are not limited in the embodiments of this application.
4) The edge cloud gateway 210 is mainly configured for intelligent scheduling, load balancing, statistical analysis, and the like of local traffic, that is, traffic within a geographical range for which the edge cloud gateway 210 is responsible, to schedule service traffic to a specific MEC processing server. In an embodiment of this application, when an edge service is deployed, the edge cloud gateway 210 also needs to be pre-configured. Specifically, the edge cloud gateway 210 corresponds to the second pre-configuration rule, and the second pre-configuration rule includes at least a preset offloading policy.
For example, the preset offloading policy is to perform offloading based on IP 5-tuple information. For example, usually, an IP 5-tuple includes a source IP address, a destination IP address, a protocol number, a source port, and a destination port, and the offloading can be performed based on a load balancing algorithm. For example, the load balancing algorithm is source address hashing. The source IP address is hashed, and a hash value is obtained through the hash calculation. Based on the hash value and configuration rules of a MEC processing server list, a corresponding MEC processing server is determined, and further, the service request is forwarded to the corresponding MEC processing server for processing.
In another example, the preset offloading policy is to perform offloading based on uniform resource locator (URL) information. For example, the service request may be forwarded to a MEC processing server that processes a corresponding service type according to a service type of the domain name request in the service request, and a specific domain name or resource type stored on each MEC processing server, for example, a specific domain name or resource type representing a service type of processing an image or video.
The specific offloading policy is not limited in this embodiment of this application, and may be selected and used according to actual requirements.
It should be noted that the second pre-configuration rule is configured by the configured edge controller 230, or is configured locally. That is, the edge cloud gateway 210 not only can be configured and managed by the edge controller 230 in a unified manner, but also can be configured and loaded locally, which is not limited in this embodiment of this application.
Specifically, the edge cloud gateway 210 is configured to perform the following steps:

S1: Receive a first HTTP service request forwarded by the UPF 200, the first HTTP service request being forwarded by the UPF 200 upon parsing that a destination address in the first HTTP service request transmitted by the terminal is an edge-application VIP.
S2: Determine a corresponding MEC processing server according to the first HTTP service request and a preset offloading policy, and offload the first HTTP service request to the corresponding MEC processing server.

Further, for the standard DNS requesting process and the HTTP DNS requesting process, the following two implementations may be performed before the edge cloud gateways performs step S1.
First implementation: For a standard DNS requesting process, that is, for a case in which the terminal initiates a DNS request for an original domain name, the edge cloud gateway 210 is further configured to perform the following steps:

1) Receive a second HTTP service request forwarded by the UPF 200, where the second HTTP service request is forwarded by the UPF 200 upon parsing that a destination address in the second HTTP service request transmitted by the terminal is the global edge gateway IP address, the global edge gateway IP address is returned to the terminal by the GSLB 220 upon receiving a DNS request transmitted by the terminal, and the global edge gateway IP address is an IP address that identifies service acceleration.
2) Obtain the edge-application VIP from the GSLB 220 based on the second HTTP service request.

Specifically, the following steps are performed:

a. Transmit a first HTTP DNS request to the GSLB 220, the first HTTP DNS request including at least a service acceleration domain name and a source address, the source address being an IP address of the edge cloud gateway 210.
b. Receive a first HTTP DNS response returned by the GSLB 220, the first HTTP DNS response including the edge-application VIP, the edge-application VIP being determined by the GSLB 220 according to the IP address of the edge cloud gateway 210 and the service acceleration domain name in the first HTTP DNS request.

3) Transmit a redirection response to the terminal, a redirection address in the redirection response being the edge-application VIP, so that the terminal transmits the first HTTP service request based on the redirection response, that is, the first HTTP service request in S1.
The redirection response is an HTTP 302 redirection response, and is used for causing the terminal to transmit the first HTTP service request.
Second implementation: For an HTTP DNS requesting process, that is, for a case in which the terminal initiates an HTTP DNS request for an original domain name,
the first HTTP service request is transmitted by the terminal upon receiving a second HTTP DNS response forwarded by the UPF 200, the second HTTP DNS response is returned by the GSLB 220 upon receiving a second HTTP DNS request forwarded by the UPF 200, and the second HTTP DNS request is forwarded by the UPF 200 upon parsing that a destination address in the second HTTP DNS request transmitted by the terminal is an IP address of the GSLB 220. The second HTTP DNS response includes at least an edge-application VIP, the edge-application VIP being determined by the GSLB 220 according to a service acceleration domain name in the second HTTP DNS request.
In this way, in this embodiment of this application, for the conventional DNS addressing process or HTTP DNS requesting process, an edge cloud gateway and an edge controller are added, so that a terminal can initiate a DNS request or an HTTP DNS request for an original domain name by using a standard DNS request or an HTTP DNS request. For a DNS request of the terminal, a global edge gateway IP address that identifies service acceleration is returned for a domain name that requires service acceleration, and further, processing can be performed by using the edge cloud gateway, to obtain an edge-application VIP from the GSLB, so that the terminal retransmits a service request based on the edge-application VIP, and the service request of the terminal is offloaded to a specific MEC processing server. Alternatively, for an HTTP DNS request of the terminal, the terminal obtains the edge-application VIP from the GSLB by using the UPF, and further may initiate a first HTTP service request based on the edge-application VIP, and forward the first HTTP service request to the edge cloud gateway, and the edge cloud gateway processes the first HTTP service request, and offloads the first HTTP service request to a specific MEC processing server. Because the edge cloud gateway is closer to the terminal and the edge cloud gateway can convert the original DNS request into an HTTP DNS request to bypass a local DNS so as to simplify the access procedure, processing rate is higher, latency is reduced, and a nearest edge computing node can be selected, thereby achieving purpose of edge acceleration. In addition, various over-the-top (OTT) services can be provided for users through the Internet in the framework of the related art, and an MEC node can be smoothly accessed without modification, which is not perceived by the users, thereby improving the ease of implementation and user experience.
Based on the architectural diagram of the system shown in FIG. 2 in the foregoing embodiment, FIG. 3 is a flowchart of a method for selecting an MEC node according to an embodiment of this application, which is mainly applied to an edge cloud gateway. The method includes the following steps.
Step 300: The edge cloud gateway receives a first HTTP service request forwarded by a UPF, a destination address of the first HTTP service request being an edge-application VIP.
Step 310: Determine a corresponding MEC processing server according to the first HTTP service request and a preset offloading policy, and offload the first HTTP service request to the corresponding MEC processing server.
Before step 300 is performed, the edge-application VIP may be obtained in different manners according to different request statuses of the terminal, and procedures for triggering step 300 are different. Specifically, the following two manners may be adopted.
First manner: For a standard DNS requesting process, before the edge cloud gateway receives the first HTTP service request forwarded by the UPF, the following steps are further included:

1) Receive a second HTTP service request forwarded by the UPF, a destination address of the second HTTP service request being a global edge gateway IP address, the global edge gateway IP address being returned to the terminal by the GSLB upon receiving a DNS request transmitted by the terminal, the global edge gateway IP address being an IP address that identifies service acceleration.
2) Obtain the edge-application VIP from the GSLB based on the second HTTP service request.

Specifically, the following steps are included:

(1) Transmit a first HTTP DNS request to the GSLB, the first HTTP DNS request including at least a service acceleration domain name and a source address, the source address being an IP address of the edge cloud gateway.
(2) Receive a first HTTP DNS response returned by the GSLB, the first HTTP DNS response including the edge-application VIP, the edge-application VIP being determined by the GSLB according to the IP address of the edge cloud gateway and the service acceleration domain name in the first HTTP DNS request.
(3) Transmit a redirection response to the terminal, a redirection address in the redirection response being the edge-application VIP, so that the terminal transmits the first HTTP service request based on the redirection response.

It should be noted that for the first manner, during specific execution, the UPF performs a corresponding action according to a first pre-configuration rule thereof, the GSLB performs a corresponding action according to a third pre-configuration rule thereof, and the edge cloud gateway performs a corresponding action according to a second pre-configuration rule thereof, so that the edge cloud gateway converts the DNS request of the terminal into the first HTTP DNS request so as to simplify the access process and reduce latency. In addition, the first HTTP service request of the terminal is forwarded to the edge cloud gateway, so that a nearest edge computing node is selected, thereby achieving purpose of service acceleration.
Second manner: For an HTTP DNS requesting process, before the edge cloud gateway receives the first HTTP service request forwarded by the UPF, the first HTTP service request is transmitted by the terminal upon receiving a second HTTP DNS response forwarded by the UPF, the second HTTP DNS response being returned by the GSLB upon receiving a second HTTP DNS request forwarded by the UPF, the second HTTP DNS request being forwarded by the UPF upon parsing that a destination address in the second HTTP DNS request transmitted by the terminal is an IP address of the GSLB. The second HTTP DNS response includes at least an edge-application VIP, the edge-application VIP being determined by the GSLB according to a service acceleration domain name in the second HTTP DNS request.
It should be noted that for the second manner, during specific execution, the UPF performs a corresponding action according to a fourth pre-configuration rule thereof, the GSLB performs a corresponding action according to a fifth pre-configuration rule thereof, and the edge cloud gateway performs a corresponding action according to the second pre-configuration rule thereof, so that the terminal transmits the HTTP DNS request to the GSLB, obtains the edge-application VIP, transmits the first HTTP service request based on the edge-application VIP, and forwards the first HTTP service request to the edge cloud gateway. The edge cloud gateway processes the first HTTP service request and forwards the first HTTP service request to the corresponding MEC processing server, so that a nearest edge computing node is selected, thereby achieving purpose of service acceleration.
Based on the foregoing embodiments, specific application scenarios are used below for description. For different request statuses of different terminals, two different implementations may be adopted to illustrate the method for selecting an MEC node in the embodiments of this application.
First implementation: For a standard DNS requesting process, that is, for a case in which the terminal initiates a DNS request for an original domain name, the UPF corresponds to the first pre-configuration rule, the GSLB corresponds to the third pre-configuration rule, and the edge cloud gateway corresponds to the second pre-configuration rule.
FIG. 4 is an interactive flowchart of a method for selecting an MEC node according to an embodiment of this application.
Step 400: A terminal transmits a DNS request to a GSLB by using a local DNS.
Specifically, the terminal may initiate a request for an original domain name by a standard DNS request. Based on a conventional DNS addressing mechanism, the request is forwarded to the GSLB by using the local DNS.
Step 401: The GSLB receives the DNS request transmitted by the terminal.
Step 402: The GSLB returns a global edge gateway IP address to the terminal in response to determining that a domain name in the DNS request is a preset service acceleration domain name.
In this embodiment of this application, it is mainly based on the pre-configuration on the GSLB that the GSLB can implement this function. The GSLB returns the global edge gateway IP address as a destination IP according to a corresponding third pre-configuration rule.
Step 403: The terminal transmits a second HTTP service request to a UPF.
After receiving the global edge gateway IP address, the terminal initiates a second HTTP service request, where a destination address in the second HTTP service request is the global edge gateway IP address.
Step 404: The UPF receives the second HTTP service request transmitted by the terminal.
Step 405: The UPF forwards the second HTTP service request to an edge cloud gateway in response to parsing that the destination address in the second HTTP service request is the global edge gateway IP address.
This step is implemented by the UPF based on a first pre-configuration rule corresponding thereto, which includes forwarding an HTTP service request of which a destination address is the global edge gateway IP address to the edge cloud gateway.
Step 406: The edge cloud gateway receives the second HTTP service request forwarded by the UPF.
Step 407: The edge cloud gateway transmits a first HTTP DNS request to the GSLB.
The first HTTP DNS request includes at least a service acceleration domain name and a source address, the source address being an IP address of the edge cloud gateway. That is, the edge cloud gateway initiates a first HTTP DNS request to the GSLB by using the IP address thereof as the source address. In this way, the edge cloud gateway can convert the original DNS request into an HTTP DNS request, which during retransmission thereof, can be directly transmitted to the GSLB bypassing the local DNS, which simplifies the access procedure.
Step 408: The GSLB receives the first HTTP DNS request transmitted by the edge cloud gateway.
Step 409: The GSLB returns a first HTTP DNS response to the edge cloud gateway.
The first HTTP DNS response includes an edge-application VIP, the edge-application VIP being determined by the GSLB according to the IP address of the edge cloud gateway and the service acceleration domain name.
In this embodiment of this application, if the GSLB determines that the source address in the first HTTP DNS request is the IP address of the edge cloud gateway, and the requested domain name is the service acceleration domain name, an actual edge-application VIP can be selected and returned according to the IP address of the edge cloud gateway and a service configuration corresponding to the service acceleration domain name.
Step 410: The edge cloud gateway receives the first HTTP DNS response returned by the GSLB.
Step 411: The edge cloud gateway transmits a redirection response to the terminal based on the first HTTP DNS response.
A redirection address in the redirection response is the edge-application VIP.
The redirection response is an HTTP 302 redirection response, which instructs the terminal to re-initiate a service request.
Step 412: The terminal receives the redirection response transmitted by the edge cloud gateway.
Step 413: The terminal transmits a first HTTP service request to the UPF based on the redirection response.
Specifically, after receiving the redirection response, the terminal can re-initiate the service request according to standard HTTP actions, where a destination address in the first HTTP service request is the edge-application VIP.
Step 414: The UPF receives the first HTTP service request transmitted by the terminal.
Step 415: The UPF forwards the first HTTP service request to the edge cloud gateway in response to determining that the destination address of the first HTTP service request is the edge-application VIP.
This step is implemented by the UPF based on the corresponding first pre-configuration rule, which includes forwarding an HTTP service request of which a destination address is the edge-application VIP to the edge cloud gateway.
Step 416: The edge cloud gateway receives the first HTTP service request forwarded by the UPF, determines a corresponding MEC processing server according to the first HTTP service request and a preset offloading policy, and offloads the first HTTP service request to the corresponding MEC processing server.
In this way, in this embodiment of this application, a service data procedure for selecting an MEC node based on a standard DNS requesting process is provided mainly for a 5G mobile communication network. Devices, such as the edge cloud gateway, may offload a service to the nearest MEC processing server with higher accuracy so as to select an MEC node and reduce latency, thereby achieving purpose of service acceleration.
Second implementation: For an HTTP DNS requesting process, that is, for a case in which the terminal initiates an HTTP DNS request for an original domain name, the UPF corresponds to the fourth pre-configuration rule, the GSLB corresponds to the fifth pre-configuration rule, and the edge cloud gateway corresponds to the second pre-configuration rule.
FIG. 5 is an interactive flowchart of a method for selecting an MEC node according to an embodiment of this application.
Step 500: A terminal transmits a second HTTP DNS request to a UPF.
Specifically, the terminal can initiate a request for an original domain name by using an HTTP DNS, that is, the terminal may be based on an HTTP DNS addressing mechanism.
A destination address in the second HTTP DNS request is an IP address of a GSLB.
Step 501: The UPF receives the second HTTP DNS request transmitted by the terminal.
Step 502: The UPF forwards the second HTTP DNS request to the GSLB in response to determining that the destination address in the second HTTP DNS request is the IP address of the GSLB.
Step 503: The GSLB receives the second HTTP DNS request forwarded by the UPF.
Step 504: The GSLB returns a second HTTP DNS response to the UPF.
The second HTTP DNS response includes at least an edge-application VIP.
Step 505: The UPF receives the second HTTP DNS response returned by the GSLB.
Step 506: The UPF forwards the second HTTP DNS response to the terminal.
Step 507: The terminal receives the second HTTP DNS response forwarded by the UPF.
Step 508: The terminal transmits a first HTTP service request to the UPF based on the edge-application VIP in the second HTTP DNS response.
A destination address of the first HTTP service request is the edge-application VIP.
Step 509: The UPF receives the first HTTP service request transmitted by the terminal.
Step 510: The UPF forwards the first HTTP service request to the edge cloud gateway in response to determining that the destination address of the first HTTP service request is the edge-application VIP.
Step 511: The edge cloud gateway receives the first HTTP service request forwarded by the UPF, determines a corresponding MEC processing server according to the first HTTP service request and a preset offloading policy, and offloads the first HTTP service request to the corresponding MEC processing server.
In this way, in this embodiment of this application, another service data procedure for selecting an MEC node based on an HTTP DNS requesting process is provided mainly for a 5G mobile communication network. Devices, such as the edge cloud gateway, may offload a service to the nearest MEC processing server with higher accuracy so as to select an MEC node and reduce latency, thereby achieving purpose of service acceleration.
Based on the foregoing embodiments, FIG. 6 shows an edge cloud gateway for selecting an MEC node according to an embodiment of this application. The edge cloud gateway is configured on an MEC node side or an edge DC side in a system for selecting an MEC node. The system includes at least a UPF and the edge cloud gateway, and the edge cloud gateway includes:

a first receiving module 60, configured to receive a first HTTP service request forwarded by the UPF, a destination address of the first HTTP service request being an edge-application VIP;
a determining module 61, configured to determine a corresponding MEC processing server according to the first HTTP service request and a preset offloading policy; and
an offloading module 62, configured to offload the first HTTP service request to the corresponding MEC processing server.

In an embodiment, the system further includes a GSLB, and the edge cloud gateway further includes:

a second receiving module 63, configured to receive a second HTTP service request forwarded by the UPF, a destination address of the second HTTP service request being a global edge gateway IP address, the global edge gateway IP address being returned to a terminal by the GSLB upon receiving a DNS request transmitted by the terminal, the global edge gateway IP address being an IP address that identifies service acceleration;
an obtaining module 64, configured to obtain the edge-application VIP from the GSLB based on the second HTTP service request; and

module

In an embodiment, the obtaining module 64 is specifically configured to:

transmit a first HTTP DNS request to the GSLB, the first HTTP DNS request including at least a service acceleration domain name and a source address, the source address being an IP address of the edge cloud gateway; and
receive a first HTTP DNS response returned by the GSLB, the first HTTP DNS response including the edge-application VIP, the edge-application VIP being determined by the GSLB according to the IP address of the edge cloud gateway and the service acceleration domain name in the first HTTP DNS request.

In an embodiment, the redirection response is an HTTP 302 redirection response.
In an embodiment, the first HTTP service request is transmitted by the terminal upon receiving a second HTTP DNS response forwarded by the UPF, the second HTTP DNS response being returned by the GSLB upon receiving a second HTTP DNS request forwarded by the UPF, the second HTTP DNS request being forwarded by the UPF upon parsing that a destination address in the second HTTP DNS request transmitted by the terminal is an IP address of the GSLB.
The second HTTP DNS response includes at least the edge-application VIP, the edge-application VIP being determined by the GSLB according to a service acceleration domain name in the second HTTP DNS request.
In an embodiment, the UPF, the edge cloud gateway, and the GSLB respectively correspond to different pre-configuration rules, the pre-configuration rule representing a routing configuration for service acceleration.
In an embodiment, the UPF corresponds to a first pre-configuration rule, the first pre-configuration rule including at least: forwarding an HTTP service request of which a destination address is the global edge gateway IP address to the edge cloud gateway; or forwarding an HTTP service request of which a destination address is the edge-application VIP to the edge cloud gateway.
The edge cloud gateway corresponds to a second pre-configuration rule, the second pre-configuration rule including at least: the preset offloading policy.
The GSLB corresponds to a third pre-configuration rule, the third pre-configuration rule including at least: for a DNS request of which a domain name is a preset service acceleration domain name, returning the global edge gateway IP address; or for an HTTP DNS request of which a domain name is a preset service acceleration domain name, returning a corresponding edge-application VIP based on a source address and the service acceleration domain name in the HTTP DNS request, the source address being the IP address of the edge cloud gateway.
In an embodiment, the UPF corresponds to a fourth pre-configuration rule, the fourth pre-configuration rule including at least: forwarding an HTTP service request of which a destination address is the edge-application VIP to the edge cloud gateway.
The GSLB corresponds to a fifth pre-configuration rule, the fifth pre-configuration rule including at least: for an HTTP DNS request of which a domain name is a preset service acceleration domain name, returning a corresponding edge-application VIP based on the service acceleration domain name in the HTTP DNS request.
In an embodiment, the first pre-configuration rule, the second pre-configuration rule, or the fourth pre-configuration rule is configured by a configured edge controller or is locally configured.
The edge controller is configured in a cloud center of a core network in the system, and communicably connected to the edge cloud gateway and the UPF.
Based on the foregoing embodiments, FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of this application.
This embodiment of this application provides an electronic device. The electronic device may include a center processing unit (CPU) 710, a memory 720, an input device 730, an output device 740, and the like. The input device 730 may include a keyboard, a mouse, a touch screen, and the like. The output device 740 may include a display device, such as a liquid crystal display (LCD) or a cathode ray tube (CRT).
The memory 720 may include a read-only memory (ROM) and a random access memory (RAM), and provide program instructions and data stored in the memory 720 for the processor 710. In this embodiment of this application, the memory 720 may be configured to store a program of the method for selecting an MEC node according to any one of the embodiments of this application.
The processor 710 invokes the program instructions stored in the memory 720, and is configured to perform, according to the obtained program instructions, the method for selecting an MEC node according to any one of the embodiments of this application.
Based on the foregoing embodiments, an embodiment of this application provides a computer-readable storage medium, storing a computer program, the computer program, when executed by a processor, implementing the method for selecting an MEC node according to any one of the method embodiments described above.
A person skilled in the art can understand that the embodiments of this application may be provided as a method, a system, or a computer program product. Therefore, this application may use a form of hardware-only embodiments, software-only embodiments, or embodiments combining software and hardware. Moreover, this application may use a form of a computer program product that is implemented on one or more computer-usable storage media (including but not limited to a disk memory, a CD-ROM, an optical memory, and the like) that include computer-usable program code.
In an exemplary embodiment, a computer program product is further provided, when executed, the computer program product is used to implement the method for selecting an MEC node provided in the foregoing embodiments.
This application is described with reference to flowcharts and/or block diagrams of the method, the device (system), and the computer program product according to the embodiments of this application. It is to be understood that computer program instructions can implement each procedure and/or block in the flowcharts and/or block diagrams and a combination of procedures and/or blocks in the flowcharts and/or block diagrams. The computer program instructions may be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or a processor of another programmable data processing device to generate a machine, so that an apparatus configured to implement functions specified in one or more procedures in the flowcharts and/or one or more blocks in the block diagrams is generated by using instructions executed by the computer or the processor of another programmable data processing device.
The computer program instructions may be alternatively stored in a computer-readable memory that can instruct a computer or another programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate an artifact that includes an instruction apparatus. The instruction apparatus implements a specific function in one or more procedures in the flowcharts and/or in one or more blocks in the block diagrams.
The computer program instructions may further be loaded onto a computer or another programmable data processing device, so that a series of operations and steps are performed on the computer or the another programmable device, thereby generating computer-implemented processing. Therefore, the instructions executed on the computer or the another programmable device provide steps for implementing a specific function in one or more processes in the flowcharts and/or in one or more blocks in the block diagrams.
Although the invention has been explained in relation to its preferred embodiment(s) as mentioned above, it is to be understood that many other possible modifications and variations can be made without departing from the scope of the present invention, which is defined by the claims.

Claims

A method for selecting a mobile edge computing, MEC, node, performed by an edge cloud gateway (210), the edge cloud gateway (210) being configured in a system for selecting an MEC node, the system comprising at least a user plane function, UPF (200), and the edge cloud gateway (210), the method comprising:
receiving (300), by the edge cloud gateway (210), a first HyperText Transfer Protocol, HTTP, service request forwarded by the UPF (200), a destination address of the first HTTP service request being an edge-application virtual Internet Protocol address, VIP;

determining (310) a corresponding MEC processing server according to the first HTTP service request and a preset offloading policy; and

offloading the first HTTP service request to the corresponding MEC processing server,

wherein the system further comprises a global server load balance, GSLB (220), and before the receiving (300), by the edge cloud gateway (210), the first HTTP service request
forwarded by the UPF (200), the method further comprises:
receiving a second HTTP service request forwarded by the UPF (200), a destination address of the second HTTP service request being a global edge gateway IP address, the global edge gateway IP address being returned to a terminal by the GSLB (220) upon receiving a domain name system, DNS, request transmitted by the terminal, the global edge gateway IP address being an IP address that identifies service acceleration;

obtaining the edge-application VIP from the GSLB (220) based on the second HTTP service request; and

transmitting a redirection response to the terminal, a redirection address in the redirection response being the edge-application VIP, whereby the terminal transmits the first HTTP service request based on the redirection response.
The method according to claim 1, wherein the obtaining the edge-application VIP from the GSLB (220) comprises:
transmitting a first HTTP DNS request to the GSLB (220), the first HTTP DNS request comprising at least a service acceleration domain name and a source address, the source address being an IP address of the edge cloud gateway (210); and

receiving a first HTTP DNS response returned by the GSLB (220), the first HTTP DNS response comprising the edge-application VIP, the edge-application VIP being determined by the GSLB (220) according to the IP address of the edge cloud gateway (210) and the service acceleration domain name in the first HTTP DNS request.
The method according to claim 1, wherein the redirection response is an HTTP 302 redirection response.
The method according to claim 1, wherein the first HTTP service request is transmitted by a terminal upon receiving a second HTTP DNS response forwarded by the UPF (200), the second HTTP DNS response being returned by the GSLB (220) upon receiving a second HTTP DNS request forwarded by the UPF (200), the second HTTP DNS request being forwarded by the UPF (200) upon parsing that a destination address in the second HTTP DNS request transmitted by the terminal is an IP address of the GSLB (220); and
the second HTTP DNS response comprises at least the edge-application VIP, the edge-application VIP being determined by the GSLB (220) according to a service acceleration domain name in the second HTTP DNS request.
The method according to any one of claims 1 to 4, wherein the UPF (200), the edge cloud gateway (210), and the GSLB (220) respectively correspond to different pre-configuration rules, the pre-configuration rule representing a routing configuration for service acceleration.
The method according to claim 5, wherein the UPF (200) corresponds to a first pre-configuration rule, the first pre-configuration rule comprising at least: forwarding an HTTP service request of which a destination address is the global edge gateway IP address to the edge cloud gateway (210); or forwarding an HTTP service request of which a destination address is the edge-application VIP to the edge cloud gateway (210);
the edge cloud gateway (210) corresponds to a second pre-configuration rule, the second pre-configuration rule comprising at least: the preset offloading policy; and

the GSLB (220) corresponds to a third pre-configuration rule, the third pre-configuration rule comprising at least: for a DNS request of which a domain name is a preset service acceleration domain name, returning the global edge gateway IP address; or for an HTTP DNS request of which a domain name is a preset service acceleration domain name, returning a corresponding edge-application VIP based on a source address and the service acceleration domain name in the HTTP DNS request, the source address being the IP address of the edge cloud gateway (210).
The method according to claim 5, wherein the UPF (200) corresponds to a fourth pre-configuration rule, the fourth pre-configuration rule comprising at least: forwarding an HTTP service request of which a destination address is the edge-application VIP to the edge cloud gateway (210); and
the GSLB (220) corresponds to a fifth pre-configuration rule, the fifth pre-configuration rule comprising at least: for an HTTP DNS request of which a domain name is a preset service acceleration domain name, returning a corresponding edge-application VIP based on the service acceleration domain name in the HTTP DNS request.
The method according to claim 6 or 7, wherein the first pre-configuration rule, the second pre-configuration rule, or the fourth pre-configuration rule is configured by a configured edge controller or (230) is locally configured,
the edge controller (230) being configured in a cloud center of a core network in the system, and being communicably connected to the edge cloud gateway (210) and the UPF (200).
The method according to any one of claims 1 to 4, wherein the edge cloud gateway (210) is configured on an MEC node side or an edge data center, DC, side of the system.
A system for selecting a mobile edge computing, MEC ,node, comprising at least a user plane function, UPF (200), and an edge cloud gateway (210),
the UPF (200) being configured to receive a first HyperText Transfer Protocol, HTTP, service request transmitted by a terminal, and forward the first HTTP service request to the edge cloud gateway (210) in response to determining that a destination address in the first HTTP service request is an edge-application virtual Internet Protocol address, VIP; and

the edge cloud gateway (210) being configured to receive the first HTTP service request forwarded by the UPF (200), determine a corresponding MEC processing server based on the first HTTP service request and a preset offloading policy, and offload the first HTTP service request to the corresponding MEC processing server,

wherein the system further comprises a global server load balance, GSLB (220),

and wherein, before the receiving by the edge cloud gateway (210) the first HTTP service request forwarded by the UPF (200):
the GSLB (220) is configured to receive a domain name system, DNS, request transmitted by the terminal, and return a global edge gateway IP address to the terminal in response to determining that a domain name in the DNS request is a preset service acceleration domain name, the global edge gateway IP address being an IP address that identifies service acceleration;

the UPF (200) is further configured to receive a second HTTP service request transmitted by the terminal, and forward the second HTTP service request to the edge cloud gateway (210) in response to determining that a destination address in the second HTTP service request is the global edge gateway IP address; and

the edge cloud gateway (210) is further configured to receive the second HTTP service request forwarded by the UPF (200), obtain an edge-application VIP from the GSLB (220) based on the second HTTP service request, and transmit a redirection response to the terminal, a redirection address in the redirection response being the edge-application VIP, whereby the terminal is configured to transmit the first HTTP service request based on the redirection response.
An edge cloud gateway, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor, when executing the program, performing the method according to any one of claims 1 to 9.
A computer program product comprising instructions, the instructions, when run on a computer, causing the computer to perform the method according to any one of claims 1 to 9.
A storage medium, configured to store the computer program product of claim 12.